ML Observability Insights
Amber and Xander discuss the automation of ML monitoring through observability, emphasizing how it helps catch production issues early. They explore different types of drift, such as feature and label drift, and highlight the capabilities of Arize's Phoenix library in comparing embeddings. Additionally, they address the importance of monitoring for bias by analyzing how models treat sensitive groups.In this clip
From this podcast
Super Data Science: ML & AI Podcast with Jon Krohn
689: Observing LLMs in Production to Automatically Catch Issues — with Amber Roberts and Xander Song
Related Questions